Score Test for Homogeneity of Variances in Normal Distributions

Abstract

In this study, we suggest a novel test statistic based on the Score statistic for evaluating the homogeneity of variances in normal distributions. In addition to the conventional chi-square approximation of the Score statistic, we introduce a parametric bootstrap technique known as the Computational Approach Test (CAT). Through a simulation study, we evaluate the proposed test’s CAT approach (referred to as CS) and assess its performance against established methods under varying group sizes and sample sizes. The results show that, regardless of the number of groups, the CAT approach of the Score test performs well when sample sizes and variances are directly proportional, even with a minimum sample size of three. Furthermore, when sample sizes and variances are inversely proportional, the proposed test significantly outperforms alternative methods. To demonstrate the application of the discussed methods, we provide two numerical examples.

Keywords

• Homogeneity of variances
• Computational approach test
• Score test

References

1. [1] Boos, D. D., Brownie, C., “Comparing variances and other measures of dispersion”, Statistical Science, 19: 571-578, (2004). DOI: https://doi.org/10.1214/088342304000000500.
2. [2] Li, X., Qiu, W., Morrow, J., DeMeo, D. L., Weiss, S. T., Fu, Y., Wang, X., “A comparative study of tests for homogeneity of variances with application to DNA methylation data”, PloS one, 10(12): (2015). DOI: https://doi.org/10.1371/journal.pone.0145295
3. [3] Conover, W. J., Johnson, M.E., Johnson, M.M. “A comparative study of tests for homogeneity of variances, with applications to the outer continental shelf bidding data”, Technometrics, 23: 351- 361, (1981). DOI: https://doi.org/10.1080/00401706.1981.10487693
4. [4] Cahoy, D. O. “A bootstrap test for equality of variances”, Computational Statistics and Data Analysis, 54: 2306-2316, (2010). DOI: https://doi.org/10.1016/j.csda.2010.01.015
5. [5] Bartlett, M. S., “Properties of Sufficiency and Statistical Test”, Proceedings of the Royal Society: A, 160: 268-282, (1937). DOI: https://doi.org/10.1098/rspa.1937.0109
6. [6] Bishop, D. J., Nair, U. S. “A note on certain methods of testing for the homogeneity of a set of estimated variances”, Journal of the Royal Statistical Society, 6: 89–99, (1939).
7. [7] Box, G. E., “Non-Normality and Tests on Variances”, Biometrika, 40: 318-335, (1953). DOI: https://doi.org/10.1093/biomet/40.3-4.318
8. [8] Brown, M. B., Forsythe, A.B., “Robust tests for the equality of variances”, Journal of the American Statistical Association, 69: 364-367, (1974). DOI: https://doi.org/10.2307/2285659

9. [9] Cochran, W. G., “Testing a linear relation among variances”, Biometrics, 7: 17-32, (1951). DOI: https://doi.org/10.2307/3001666
10. [10] Levene, H., “Robust Tests for Equality of Variances, In Contributions to Probability and Statistics: Essays in honor of Harold Hotelling”, 2: 278-292, (1960).
11. [11] Loh, W. Y., “Some modifications of Levene's test of variance homogeneity”, Journal of Statistical Computation and Simulation, 28: 213-226, (1987). DOI: https://doi.org/10.1080/00949658708811047
12. [12] Keyes, T. K., Levy, M. S., “Analysis of Levene’s Test under Design Inbalance”, Journal of Educational and Behavioral Statistics, 22: 227-236, (1997). DOI: https://doi.org/10.3102/10769986022003227
13. [13] Bhandary, M., Dai, H., “An alternative test for the equality of variances for several populations when the underlying distributions are normal”, Communications in Statistics- -Simulation and Computation, 38: 109-117, (2008). DOI: https://doi.org/10.1080/03610910802387378
14. [14] Liu, X., Xu, X., “A new generalized p-value approach for testing the homogeneity of variances”, Statistics and Probability Letters, 80: 1486-1491, (2010). DOI: https://doi.org/10.1016/j.spl.2010.04.018
15. [15] Gökpınar, E., Gökpınar, F., “Testing equality of variances for several normal populations”, Communications in Statistics- Simulation and Computation, 46(1): 38–52, (2017). DOI: https://doi.org/10.1080/03610918.2015.1093934
16.   [16] Jafari, A. A., Shaabani, J., “Comparing scale parameters in several gamma distributions with known shapes”, Computational Statistics, 35(4): 1927-1950, (2020). DOI: https://doi.org/10.1007/s00180-019-00926-0
17. [17] Wang, J., Li, X., Liang, H., “A new exact p-value approach for testing variance homogeneity”, Statistics Theory Related Fields, 6(1): 81-86, (2022). DOI: https://doi.org/10.1080/24754269.2021.2021010
18. [18] Bera, A. K., Bilias, Y., “Rao's score, Neyman's C (α) and Silvey's LM tests: an essay on historical developments and some new results”, Journal of Statistical Planning and Inference, 97: 9-44, (2001). DOI: https://doi.org/10.1016/S0378-3758(00)00236-2
19. [19] Davison, A. C., Hinkley, D. V., “Bootstrap methods and their application”, Cambridge: Cambridge University Press, (1997). DOI: https://doi.org/10.1017/CBO9780511802843
20. [20] Chang, C. H., Pal, N., Lin, J. J., “A revisit to test the equality of variances of several populations”, Communications in Statistics- -Simulation and Computation, 46: 6360-6384, (2017). DOI: https://doi.org/10.1080/03610918.2016.1148143
21. [21] Gökpınar, E., “Standardized likelihood ratio test for homogeneity of variance of several normal populations”, Communications in Statistics- -Simulation and Computation, 51: 6309-6319, (2022). DOI: https://doi.org/10.1080/03610918.2021.1943494
22. [22] Anderson, E., “The Irises of the Gaspe Peninsula”, Bulletin of the American Iris Society, 59: 2–5, (1935).
23. [23] Hu, Y. C., “A new fuzzy-data mining method for pattern classification by principal component analysis”, Cybernetics and Systems, 36: 527–547, (2005). DOI: https://doi.org/ 10.1080/01969720590913116
24. [24] Korkmaz, S., Goksuluk, D., Zararsiz, G., “MVN: An R Package for Assessing Multivariate Normality”, The R Journal, 6: 151–162, (2014). DOI: https://doi.org/10.32614/RJ-2014-031